December 2023 Summaries

Harness is transitioning from its CD Community Edition to Gitness, an open-source software development platform launched in September 2023, which offers a comprehensive solution for continuous integration and delivery (CI/CD). Initially launched in January 2022, the Harness CD Community Edition was intended to help users learn and use CD & GitOps with minimal resources and self-management. However, Gitness, which includes a Git-compliant source code management system and Drone pipelines, is seen as a more effective solution for delivering a simple CD pipeline. With Gitness, users can host git repositories and manage automated CI/CD pipelines in one platform, enhancing collaboration and streamlining development workflows. Gitness supports basic CD tasks, including application builds, Docker image creation, and Kubernetes deployments, and can be upgraded to Harness CD & GitOps for more advanced functionalities like support for various deployment strategies, governance, and intelligent deployment verification. Existing users of the Harness CD Community Edition are encouraged to transition to Gitness, as future releases of the Community Edition will be discontinued, and its source code repositories will become private.

In December 2023, Harness introduced a range of updates to enhance its product suite, focusing on improving software delivery processes and user experience. Key updates included new stage-level timeouts in Continuous Delivery & GitOps, advanced Chaos Engineering features such as a new log viewer and support for Service Mesh Performance (SMP), and expanded capabilities in the Internal Developer Portal with new scorecards and plugin support. Additional enhancements across various platforms included support for SCM PRs reports, new roles for public dashboards, and increased compatibility for Mac systems in Continuous Error Tracking. The Self-Managed Enterprise Edition received updates for secrets audit support, improved security features with domain allowlists, and the introduction of new build and push steps for Google Artifact Registry. Furthermore, Harness introduced early access features such as AIDA™ replacing Coveo for support case deflection and updates to green services in ECS deployments, alongside improvements in Software Engineering Insights with new support for mapping GitHub Actions artifacts and churn rate metrics in Jira.

Software delivery is increasingly crucial across organizations, but engineering teams face significant challenges due to the complexity and inefficiency of the current software delivery lifecycle (SDLC). Martin Reynolds, Field CTO at Harness, discusses these issues with Tim Phillips, highlighting how manual tasks and inefficiencies divert developers from their core coding responsibilities. A survey by Harness indicates that 40% of developers' time is consumed by non-coding tasks. Harness aims to streamline the SDLC by integrating processes, using AI and ML for security testing, and providing tools that enable faster and more reliable deployments. Martin shares his experience using Harness, noting its efficiency in reducing deployment times and its ability to serve as a collaborative partner in addressing software delivery challenges. By showcasing success stories from companies like Ancestry and Citi, Harness demonstrates its impact on reducing production incidents and accelerating release times, ultimately giving developers time back to focus on innovation.

In the software development landscape, balancing continuous integration with quality assurance poses a significant challenge, which ttl.sh and Gitness aim to address by creating temporary CI environments that expedite development processes. Ttl.sh is an ephemeral Docker image registry that enables temporary image tagging with built-in expiry, while Gitness, an open-source platform by Harness, simplifies the management of source code repositories and development pipelines. By leveraging these tools, developers can establish a workflow where the creation of a pull request triggers a build in Gitness, leading to Docker image creation, automated testing, and image promotion upon successful testing. This approach ensures only quality-assured images reach the central registry, reducing clutter and promoting a cleaner CI process. Additionally, the use of short-lived images and UUIDs mitigates security risks, although deploying a private version of ttl.sh might be advisable for production environments. Ultimately, integrating ttl.sh and Gitness streamlines the build and test process, avoiding outdated images and maintaining an efficient development workflow, promoting more frequent and reliable software delivery. Dewan Ahmed, a Principal Developer Advocate at Harness, brings extensive experience in solving DevOps and infrastructure challenges, advocating for underrepresented groups, and contributing to the tech community through public speaking and writing.

The Software Development Life Cycle (SDLC) is a structured methodology used to develop high-quality software, comprising seven phases: Planning, Requirements Analysis, Design, Coding, Testing, Deployment, and Maintenance. Each phase serves a crucial role, from defining the software's purpose and scope to gathering user expectations, designing architecture, writing code, testing for bugs, deploying the software, and maintaining it for continual improvement. Various SDLC models, such as Waterfall, Agile, and RAD, cater to different project needs, with the choice of model depending on factors like flexibility and timeline. Harness optimizes the SDLC process by leveraging AI to automate tasks, provide data-driven insights, and ensure efficient and scalable software delivery. The Maintenance phase involves ongoing support and upgrades to adapt to users' evolving needs, while Continuous Delivery practices enable reliable software releases through automated processes.

Implementing the Harness React SDK allows developers to easily incorporate feature flags into React applications, enabling dynamic UI changes without a full redeploy by leveraging feature toggle states. The article provides a straightforward guide on setting up and using the SDK, highlighting the integration process with core code examples. Essential requirements include having Node.js v12 or newer and React.js v16.7 or newer, with Vite recommended for project setup due to its enhanced features and performance. The implementation involves modifying two files: `App.jsx` for importing components such as `FFContextProvider` and `Logo.jsx` for utilizing the `useFeatureFlag` hook to conditionally render UI elements based on feature flag states set in the Harness UI. The guide concludes with successfully integrating the React SDK to enable feature flags in a core UI component, offering a practical starting point for developers exploring feature flagging with Harness.

Siemens Healthineers transitioned from manual processes to a modern SaaS, underscoring the importance of Site Reliability Engineering (SRE) foundations by focusing on crucial Service Level Indicators (SLIs) like availability and leveraging observability data for continuous improvement. The Harness Platform supports this journey by managing Service Level Objectives (SLOs), conducting chaos experiments, and validating deployments to enhance system reliability. In an episode of ShipTalk, Dr. Vlad Ukis, Head of R&D for Siemens Healthineers, discusses this transition and highlights the critical role of selecting key SLIs and creating baselines for measuring improvements. By doing so, Siemens Healthineers aims to improve the reliability of their medical devices and healthcare technology, which are vital in maintaining lives and quality of care in the healthcare industry. The Harness Platform offers robust features to support resilience goals through organizing SLOs, experimenting, and using observability data to ensure successful deployments.

Churn Rate is a crucial metric introduced by Harness Software Engineering Insights (SEI) to monitor and reduce sprint backlog volatility, thereby enhancing productivity and collaboration within development teams. By integrating Churn Rate into their workflow, organizations can create a more stable and predictable development environment, optimizing software delivery processes. The metric measures the amount of work added or removed from a sprint backlog, providing insights into scope changes and helping teams identify and address volatility. A high Churn Rate indicates significant changes, suggesting potential delays and poorly defined requirements, while a low Churn Rate points to stability and clarity. SEI's implementation of Churn Rate, using the Trellis Framework and over 40 third-party integrations, empowers teams to optimize workflows, foster collaboration between product managers and engineers, and ensure adaptability in evolving project requirements. This metric forms part of a broader strategy to baseline and continuously improve engineering processes, enabling teams to increase productivity and achieve sustained excellence in software development.